RND {coxsei}

Random number generator

Description

RND takes a vectorized positive R function defined on positive reals and returns a vector of n values of the random variable (survival time) with the specifed function as its hazard/intensity rate function.

Usage

RND(n, int, tol = .Machine$double.eps^0.5, epsabs = 1e-10, epsrel =
1e-10, limit = 1000)

Arguments

n

number of observations.

int

hazard rate function of the survival variable, or the intensity function of the one-event point process counting the number (0 or 1) of deaths by following a sample of the surviving suject.

tol

tolerance of the numerical error in calculating the inverse of the cumulative distribution function of the survival variable. Defaults to the square root of the machine epsilon.

epsabs

maximum absolute error to be tolerated by the integrator.

epsrel

maximum relative error to be tolerated by the integrator.

limit

maximum number of iterations permitted by the integrator.

Value

a vector of n observations of the survival variable with the supplied intensity/hazard function.

Author(s)

Feng Chen <[e-mail]>

Examples

set.seed(1)
dat <- RND(100,int=function(x)x^2)
ks.test(dat,pweibull,shape=3) # p-value = 0.6058
qqplot(dat,rweibull(100,shape=3))